MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202541112307 A) filed by New Prince Shri Bhavani College Of Engineering And Technology; Lara. J; Priyadharshini. K; Logeshwari P; and Kavitha. P, Chennai, Tamil Nadu, on Nov. 17, 2025, for 'fake news detection using machine learning.'
Inventor(s) include Lara. J; Priyadharshini. K; Logeshwari. P; and Kavitha. P.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "In today's fast-moving digital era, most of us rely heavily on the internet and social media platforms to access news and information. These platforms have made communication extremely fast and global, but at the same time, they have also become channels for the rapid spread of fake news. Fake news can mislead people, create unnecessary fear and panic, and in many cases, it is deliberately used for political, social, or financial motives. Because of its ability to reach millions in a short time, fake news is a serious challenge that needs technological solutions. To address this problem, our project proposes a Fake News Detection System that makes use of the XGBoost machine learning algorithm along with a PHP web application. This system allows users to input any news text and instantly receive a result showing whether the news is real or fake. The system is trained using two types of data. First, we use old, labeled datasets collected from Kaggle that already classify articles as fake or real. These datasets help the model learn important patterns and build a strong foundation. Second, to keep the system updated and adaptable, we integrate live data from News APIs. The combination of static historical datasets and dynamic live feeds allows the model to improve accuracy and stay relevant with changing news patterns. Before training, preprocessing of data is done, which includes cleaning the text, removing unnecessary words, converting all characters to lowercase, eliminating stop words, and applying tokenization. After cleaning, the data is converted into numerical format using methods like TF- IDF so that the machine learning algorithm can analyze it effectively. XGBoost is chosen because it is fast, powerful, and highly efficient for large datasets. It builds multiple decision trees and corrects mistakes step by step, resulting in high accuracy. During training, the model learns patterns such as word frequency, writing style, and tone, which help distinguish real from fake news. To make the system practical, a PHP web app is developed that provides a simple interface for users. They just enter news text, and the backend model returns results instantly. This project is useful because it reduces misinformation, ensures safer online communication, and creates a trustworthy digital environment."
Disclaimer: Curated by HT Syndication.
